DocumentCode
652732
Title
HapFACS: An Open Source API/Software to Generate FACS-Based Expressions for ECAs Animation and for Corpus Generation
Author
Amini, Rashied ; Lisetti, Christine
Author_Institution
Sch. of Comput. & Inf. Sci., Florida Int. Univ., Miami, FL, USA
fYear
2013
fDate
2-5 Sept. 2013
Firstpage
270
Lastpage
275
Abstract
We present HapFACS (ver. beta), a new open source software and API for generating FACS-based facial expressions on 3D virtual characters that have accompanying lip-synchronized animation abilities. HapFACS has two main usage scenarios: First, with the HapFACS software, users can generate repertoires of realistic FACS-validated facial expressions, either as static images or as videos, Second, with the accessible HapFACS API, users can animate speaking virtual characters with real-time realistic facial expressions, and embed these expressive characters in their own application(s) without any prior experience in computer graphics and modeling. We describe how HapFACS (1) provides control over 49 FACS Action Units at all levels of intensity, (2) enables the animation of faces with a single AU or a composition of AUs, activated unilaterally or bilaterally, and (3) can be applied to any supported character in the underlying 3D-character system1. Finally, we provide details of evaluation experiments we conducted with FACS-certified scorers to validate the facial expressions generated by HapFACS.
Keywords
application program interfaces; computer animation; public domain software; virtual reality; ECA animation; FACS-based expressions; HapFACS API; computer graphics; corpus generation; lip- synchronized animation abilities; open source API-software; real-time realistic facial expressions; realistic FACS-validated facial expressions; Emotion recognition; Facial animation; Gold; Registers; Software; Videos; EmFACS; FACS-based facial expression generation; Facial Action Coding System; HapFACS; Haptek;
fLanguage
English
Publisher
ieee
Conference_Titel
Affective Computing and Intelligent Interaction (ACII), 2013 Humaine Association Conference on
Conference_Location
Geneva
ISSN
2156-8103
Type
conf
DOI
10.1109/ACII.2013.51
Filename
6681442
Link To Document